Skip to contents

Capture output manifest for a completed step

Source: R/job_tracking_functions.R
capture_output_manifest.Rd

Creates a JSON manifest containing output file paths, sizes, and modification times. By default all files beneath output_dir are scanned; files can instead identify an exact job-specific subset. The manifest can be stored in the job tracking database and later used to verify that outputs remain intact.

Usage

capture_output_manifest(
  output_dir,
  recursive = TRUE,
  pattern = NULL,
  files = NULL
)

Arguments

output_dir

Root directory used to locate and verify output files.

recursive

Scan subdirectories (default TRUE)

pattern

Optional regex pattern to filter files.

files

Optional character vector of exact output files to record. When supplied, only these files are included and they must exist beneath output_dir. This is useful when several jobs share an output directory.

Value

JSON string containing the manifest, or NULL if directory doesn't exist